I'm not sure what you mean with these terms, or whether you would use the method from pemerton that you quoted or not.
I might use them in a different genre, using a system different from my own.
Since you used a "but", it sounds like you would not. Your terms seem contradictory. A purely narrative approach would have involved not die rolls or skills checks. Pem's approach struck me as combining simulation rules into the narrative, the very thing you said you prefer.
I prefer more simulationist games. Knowing that, it's easy to see that I'd prefer a more simulationist narrative game over simply a more narrative game. It brings the base back to simulation, which is more palatable to my personal tastes. That's all I was saying.
what I get from pemerton's post is:
- by default, the NPC is going to do X
- the PCs object and are going to try to prevent X
- figure out what rules/skills apply and make some die rolls to find out who wins
- based on the results, narrate a description that supports the result without contradicting established fact and maintaing plausibility and setting consistency.
Right, which is more narrative than I prefer most of the time, because it leaves room for "setting luck" to help or hurt on the check. By "setting luck", I mean the setting getting involved to save the day, and make the check make sense.
For example, if a villain gets +20 on their Streetwise check, odds are they know the best route to where they're headed. However, the GM initiates a skill challenge, letting the PCs contribute. One PC contributes a "shortcut" with a Streetwise check of his own, even though his bonus is only at +15. He succeeds, and the GM rules that he leads them down a parallel street while pursuing the villain, because he knows that today is the day of a parade, and that the villain will be funneled closer to their street, saving them precious time. While this makes sense, it's utilizing "setting luck" to avoid "hiccups" in the internal consistency of the game. It's the "as a matter of fact, now that you've rolled this high, the setting adds this element that wasn't there before." That's a little too much narrative control in my fantasy-genre game, but that's me. Tastes differ.
However, when running a game with a little more narrative control for the players (like Mutants and Masterminds 2e), I don't mind the approach. For example, when injured in M&M, you have to roll a Toughness save. At one point, one of the players was finally hit, and he was the least tough of the group. And it wasn't a hit, it was a crit. He used the remaining Hero Points (player narrative control tokens) to grab a feat that let him take a 20 on the Toughness save by spending a Hero Point, and then spent a Hero Point on it. When I described the scene, his character slipped while the enemy nearly hit him, but a different villain (who they were in the middle of saving) remote controlled his power suit into the way, which absorbed most of the impact for the player. This was very much messing with the narrative to fit the scene in a consistent way, but it's less simulationist than I like in a fantasy-genre game (personally, of course).
It just depends on the game for me. I don't think it's wrong to play with a narrative style. I think a narrative/simulationist approach is more palatable than a more narrative approach. Does this help clear things up?
As always, play what you like
